Optimization of Well Spacing and CO2

Optimization of Well Spacing and CO2
Miscible Flooding Startup Time in an

Ultradeep, High Pressure Oil Reservoir
Ahmad Hossein Zadeh
Petroleum Engineering
Submission date: July 2014
Supervisor:
Jon Kleppe, IPT
Norwegian University of Science and Technology

Department of Petroleum Engineering and Applied Geophysics
Optimization of Well Spacing and CO2

Miscible Flooding Startup Time in an
Ultradeep, High Pressure Oil Reservoir
TPG4920 Petroleum Engineering, Master Thesis
Ahmad Hossein Zadeh
Norwegian University of Science and Technology

Faculty of Engineering Science and Technology
Department of Petroleum Engineering and Applied Geophysics
Supervisor: Professor Jon Kleppe
Trondheim, July 2014
Summary
Exploitation of ultradeep, high pressure oil reservoirs is always associated with numerous
risks, challenges, and obstacles. One of the most pronounced constraints in development of
such fields is the number of wells, which is imposed by massively high cost of drilling. Low
number of wells may lead to high pressure isolation regions left after primary production
(pressure depletion) of the reservoir. Ultradeep reservoirs with high pressure are more prone
to such leftover high pressure isolation regions due to their low permeable characteristic.
These high pressure isolation regions can, subsequently, deteriorate the efficiency of
enhanced oil recovery (EOR) as the injected fluid cannot access the residual oil in these
regions. Therefore, well placement and inter-well spacing optimizations is of greater
importance in ultradeep, high pressure oil reservoirs to ensure higher ultimate oil recovery at
lower costs. Furthermore, due to high cost of development of ultradeep, high pressure oil
reservoirs, the EOR strategy and commencement time for the selected EOR strategy are very
critical.
The objective of this thesis is to demonstrate how optimization of well placement, well
spacing, and startup time for miscible CO2 flooding can enhance the incremental and ultimate
oil recoveries in an ultradeep, high pressure oil reservoir. To do this, a synthetic grid model
was made to run different simulation scenarios on it. The model was initialized with rock and
fluid properties within the range of those in the ultradeep, high pressure Wilcox formation in
the Gulf of Mexico to ensure that it mimics an ultradeep, high pressure oil reservoir.
The obtained results showed that optimization of well placement, well spacing, and
commencement time for any EOR strategy such as miscible CO2 flooding is very critical in
the course of making a Field Development Plan (FDP) for an ultradeep, high pressure oil
reservoir.
The results, discussions, and conclusions were finally used by the author to shed light on
potential further work on each of the aforementioned challenges in ultradeep, high pressure
oil reservoirs such as Wilcox formation in the Gulf of Mexico.
II
Acknowledgements
I would like to express my greatest appreciation to my supervisor, Professor Jon Kleppe, for
showing interest in the subject of my thesis and his subsequent supports and helps.
I would also like to thank Dr. Mohammad Ghasemi and Professor Curtis H. Whitson for
providing me with valuable information and hints that helped me a lot in selection of the
topic of my master thesis.
In addition, a thank you to all the administrative staff at the Department of Petroleum
Engineering and Applied Geophysics in NTNU for their kind supports and helps.
III
IV
Contents
INTRODUCTION ............................................................................................................................. 1
CHAPTER 1: LITERATURE REVIEW .................................................................................... 3
1.1
Oil and Gas Properties and Correlations .......................................................................................... 3
1.1.1 Critical and Reduced Properties ............................................................................................................ 3
1.1.2 Gas Pseudocritical and Pseudoreduced Properties ................................................................................ 3
1.1.3 Gas Deviation Factor............................................................................................................................. 4
1.1.4 Gas Density ........................................................................................................................................... 6
1.1.5 Gas Formation Volume Factor .............................................................................................................. 7
1.1.6 Gas Viscosity ........................................................................................................................................ 7
1.1.7 Solution Gas/Oil Ratio .......................................................................................................................... 8
1.1.8 Bubblepoint Pressure ............................................................................................................................ 8
1.1.9 Oil Viscosity ......................................................................................................................................... 8
1.1.10
Oil Compressibility ........................................................................................................................... 9
1.1.11
Oil Formation Volume Factor ........................................................................................................... 9
1.2
Miscibility ........................................................................................................................................ 10
1.2.1 Minimum Miscibility Pressure (MMP) ............................................................................................... 10
1.2.2 Effect of Numerical Dispersion on Miscible Gas Injection Simulation .............................................. 11
1.3
SENSOR Reservoir Simulator ......................................................................................................... 12
1.3.1 SENSOR Data File Structure .............................................................................................................. 12
1.3.2 Analytical Relative Permeability and Capillary Pressure Curves in SENSOR ................................... 12
1.4
Well Productivity Index Calculations .............................................................................................. 14
CHAPTER 2: MODEL PREPARATIONS............................................................................... 15

2.1
PVT Data ......................................................................................................................................... 15
2.2
Grid Model ....................................................................................................................................... 24
2.3
Relative Permeability and Capillary Pressure Curves .................................................................... 27
2.4
Well Data ......................................................................................................................................... 29
CHAPTER 3: OPTIMIZATION STRATEGY ........................................................................ 31

CHAPTER 4: RESULTS AND DISCUSSIONS ...................................................................... 33
4.1 MMP Estimation for CO2 Miscible Injection ......................................................................................... 33
4.2 Well Spacing and tD Optimization for 2-Spot Well Pattern ................................................................... 33
4.6 Effect of Optimized Well Pattern on Cumulative Oil Production and Incremental Oil Recovery ....... 54
CHAPTER 5: CONCLUSIONS................................................................................................... 57
CHAPTER 6: RECOMMENDATIONS FOR FUTURE RESEARCH ............................. 58
REFERENCES ................................................................................................................................ 59
APPENDIX 1 - SENSOR DATA FILE STRUCTURE .......................................................... 62
APPENDIX 2 FROM COMPOSITIONAL TO BLACK OIL .......................................... 64
APPENDIX 3 PHAZECOMP DATA FILE FOR MMP ESTIMATION ....................... 67
APPENDIX 4 SUBMITTED PAPER BASED ON THE THESIS WORK .................... 68
VI
List of Figures
Figure 1 - Standing-Katz Z-factor Chart (Whitson and Brule 2000) ..................................................................... 5
Figure 2 - Slimtube displacements used to determine MMP or MME (Petrowiki, Miscible Flooding 2013) ...... 11
Figure 3-Gas formation volume factor from the initial and final black oil PVT tables ........................................ 22
Figure 4-Oil formation volume factor from the initial and final black oil PVT tables: a reasonable match for
initial bubblepoint pressure .......................................................................................................................... 22
Figure 5-Oil viscosity factor from the initial and final black oil PVT tables: a perfect match for oil viscosity at
initial reservoir pressure ............................................................................................................................... 23
Figure 6-Solution gas/oil ratio from the initial and final black oil PVT tables ..................................................... 23
Figure 7-Gas viscosity from the initial and final black oil PVT tables ................................................................. 24
Figure 8-Final reservoir section grid model.......................................................................................................... 26
Figure 9-Analytical oil/water relative permeability curves generated by SENSOR ............................................. 28
Figure 10- Analytical oil/gas relative permeability curves generated by SENSOR ............................................. 29
Figure 11-Plan view of the final grid model ......................................................................................................... 34
Figure 12-Moving well positions to the boundaries of the model ........................................................................ 35
Figure 13-Moving well positions to the corners of the model .............................................................................. 36
Figure 14-Cumulative oil productions
........................................................................................................................ 36
Figure 15- Cumulative oil
.................................................................................................................. 37
Figure 16-Comparison between recoveries from optimum tD and those from other t s over tEOR for a 2-spot well
pattern .......................................................................................................................................................... 38
Figure 17-Effect of EOR startup time on oil production rate and oil recovery..................................................... 39
Figure 18-Uniform pressure distribution in grid block 1 of layer 1 with k=30 md at optimum t D and for optimum
well spacing ................................................................................................................................................. 40
Figure 19- Uniform pressure distribution in grid block 2 of layer 1 with k=19 md at optimum t D and for optimum
well spacing ................................................................................................................................................. 40
Figure 20- Uniform pressure distribution in grid block 3 of layer 1with k=0.3 md at optimum t D and for optimum
well spacing ................................................................................................................................................. 41
Figure 21- Comparison between recoveries from optimum t D and those from other t s over tEOR for a 3-spot well
pattern .......................................................................................................................................................... 42
Figure 23- Uniform pressure distribution in grid block 1 of layer 1 with k=30 md at optimum t D and for optimum
well spacing ................................................................................................................................................. 43
Figure 24- Uniform pressure distribution in grid block 2 of layer 1 with k=19 md at optimum t D and for
optimum well spacing .................................................................................................................................. 44
Figure 25- Uniform pressure distribution in grid block 3 of layer 1 with k=0.3 md at optimum t D and for
Figure 26-Moving well positions to the boundaries and corners of the model ..................................................... 45
.......................................................................................................................... 46
pattern .......................................................................................................................................................... 47
Figure 30- Non-uniform pressure distribution in grid block 1 of layer 1with k=30 md at optimum t D and for
Figure 31- Non-uniform pressure distribution in grid block 2 of layer 1with k=19 md at optimum t D a d for
VII
Figure 32- Non-uniform pressure distribution in grid block 3 of layer 1 with k=0.3 md at optimum t D and for
Figure 33- Moving well positions to the boundaries and corners of the model .................................................... 50
................................................................................................................ 50
pattern .......................................................................................................................................................... 51
Figure 36- Effect of EOR startup time on oil production rate and oil recovery.................................................... 52
Figure 37-Non-uniform pressure distribution in grid block 1 of layer 1with k=30 md at optimum t D and for
Figure 38-Non-uniform pressure distribution in grid block 2 of layer 1 with k=19 md at optimum t D and for
Figure 39-Non-uniform pressure distribution in grid block 3 of layer 1 with k=0.3 md at optimum t D and for
Figure 40- Effect of optimized well pattern on cumulative oil production ........................................................... 55
Figure 41-Effect of optimized well pattern on oil recovery .................................................................................. 55
Figure 42-Effect of optimized well pattern (number of wells) on incremental oil recovery ................................ 56
VIII
List of Tables
Table 1-Wilcox formation properties found in open literature ............................................................................. 15
Table 2-Some other properties of Wilcox formation available in open literature ................................................. 16
Table 3-Initial black oil PVT table estimated from oil and gas correlations ........................................................ 18
Table 4-Fluid composition and EOS used in the fifth SPE comparative solution project .................................... 19
Table 5-Final reservoir fluid composition and equation-of-state used for the model ........................................... 20
Table 6-Final reservoir black oil PVT table used for the model ........................................................................... 21
Table 7- Dimensions of the selected section of the cubic Wilcox reservoir model .............................................. 25
Table 8- Permeability distribution in the final grid model .................................................................................... 27
Table 9- Data used for model initialization .......................................................................................................... 27
Table 10- Variables used for analytical kr calculations in the model .................................................................... 28
Table 11-Well data in the final model .................................................................................................................. 30
Table 12-Optimum well locations and EOR startup time for 2-spot well pattern ................................................ 35
Table 13- Optimum well locations and EOR startup time for 3-spot well pattern ............................................... 41
Table 14- Optimum well locations and EOR startup time for 4-spot well pattern ............................................... 46
Table 15-Optimum well locations and EOR startup time for 5-spot well pattern ................................................ 51
Table 16-Comparison of the optimization results for different well patterns ....................................................... 54
IX
Acronyms
BHP
BML
EOR
Bottomhole Pressure
Below the Mud Line
Enhanced Oil Recovery
EOS
Equation-of-State
FDP
MME
MMP
Filed Development Plan

Minimum Miscibility Enrichment
Minimum Miscibility Pressure
OOIP
PVT
SSTVD
THP
Original Oil In Place

Pressure-Volume-Temperature
Sub-Sea True Vertical Depth
Tubing Head Pressure
XI
XII
Nomenclature
API
oil API gravity
gas specific gravity
incremental oil recovery
gas viscosity
oil viscosity
ob
saturated oil viscosity
oD
dead-oil viscosity
gas density
pr
pseudoreduced density
porosity
acentric factor
reservoir area
Bg
gas formation volume factor
Boi
initial oil formation volume factor
ci
volume shift
Co
oil compressibility
CO2
carbon dioxide
dimension
DX
length of grid blocks in X direction
DY
length of grid blocks in Y direction
reservoir thickness
well index
kr
relative permeability
krgro
relative permeability of gas at Sw=Swc and So= Sorg
krocw
relative permeability of oil at Sw=Swc and Sg= 0
krwro
relative permeability of water at Sw=1-Sorw and Sg=0
molecular weight
XIII
number of moles
ng
gas exponent for analytical kr
nog
oil/gas exponent for analytical kr
now
oil/water exponent for analytical kvolimer
nw
water exponent for analytical kr
NX
number of grid cells in X direction
NY
number of grid cells in Y direction
NZ
number of grid cells in Z direction
pressure
pb
bubblepoint pressure
pc
capillary pressure
pcr
critical pressure
ppc
pseudocritical pressure
ppr
pseudoreduced pressure
pr
reduced pressure
psat
saturation pressure
psc
standard conditions pressure
PV
pore volume
r0
rw
wellbore radius
universal gas constant
Rs
solution gas/oil ratio
Sgc
critical gas saturation
Sgr
trapped gas saturation
Sorg
residual oil saturation to gas
Sorw
residual oil saturation to water
Swc
connate water saturation
Swir
irreducible/initial water saturation
tD
EOR startup time
XIV
tEOR
simulation run duration where economic limit is reached
temperature
Tcr
critical temperature
Tpc
pseudocritical temperature
Tpr
pseudoreduced temperature
Tr
reduced temperature
Tsc
standard conditions temperature
volume
Z-factor
Zc
critical Z-factor
XV
XVI
Introduction
Outages in oil production due to political issues and declining conventional oil reservoirs in
addition to struggles to catch up with global rising oil demand have given rise to approaching
unconventional and difficult-to-develop oil resources. One type of difficult-to-develop oil
resources is ultra deepwater oil reservoirs which are extremely costly and challenging to
exploit. Some of the inherent characteristics of ultradeep oil reservoirs such as low
permeability and high initial reservoir pressure, make drilling and extraction of oil from these
reservoirs more challenging and expensive. A well-known example of such challenging
reservoirs is Wilcox sandstone in the Gulf of Mexico with a permeability of typically less
than 10 md, average oil viscosity of 5 cp, and initial reservoir pressure of about 20000 psi.
Combining these characteristics with water depth of typically greater than 5000 ft and
reservoir depth ranging from 25000 to 35000 ft below the mud line (BML) highlights the
importance of optimization process in development of Wilcox sandstone to maximize the oil
recovery at reduced costs.
Drilling wells in ultra deepwater, high pressure reservoirs has the highest contribution to
field development costs. Therefore, one of the constraints in development of such reservoirs
is the number of wells to drill. Low number of wells and improper well placement and interwell spacing in addition to structural and stratigraphic controls that compartmentalize the
reservoir can lead to localized pressure isolations at the end of primary production of high
pressure reservoirs. These leftover high pressure regions not only may contain huge amounts
of residual oil, but also can lead to high injection pressures required in the course of EOR
processes. Therefore, new dedicated wells may be required to target the accumulations in the
isolated regions and this imposes higher costs. This can be avoided by optimization of well
placement and inter-well spacing for a given number of wells to drill in ultradeep, high
pressure reservoirs in order to ensure maximized oil recovery at a minimum cost.
Furthermore, to acquire the highest incremental production for an EOR strategy, the EOR
commencement time should also be optimized.
The objective of this thesis is to demonstrate how optimization of well placement, inter-well
spacing, and EOR startup time lead to maximized ultimate and incremental oil productions in
an ultradeep, high pressure oil reservoir. The EOR process in this study is miscible CO2
1
injection and due to high-pressure characteristic of the reservoir, no secondary oil production
is considered for the reservoir model under study.
According to Muskat (1949), the well spacing problem can be approached from two
different perspectives: the physical ultimate recovery and economic ultimate recovery. The
optimization in this study was conducted based on the physical ultimate recovery obtained
from reservoir simulation. Therefore, the optimal well spacing defined in this report is only
based on the volume of oil produced and does not consider the maximum economic return
from the development of the reservoir.
Due to lack of real data, a synthetic cubic model was made and it was initialized with the
reservoir properties within the range of those of the ultradeep, high pressure Wilcox
formation in the Gulf of Mexico to ensure that the model mimics an ultradeep, high pressure
oil reservoir. The black oil table required for black oil simulations was made using the
available correlations and limited available data on Wilcox in literature. For the
compositional model, we had to make up a fluid composition and EOS which generated the
black oil table made based on the limited data available in open literature.
Inter-well spacing and EOR startup time optimization was done for four different well
patterns on the reservoir model to achieve the objective of the study. In order to reduce the
simulation cost and numerical dispersion imposed by the constraint on the number of grid
cells (6000 grid cells for SENSOR) the optimization was performed for a given well pattern
on a section of the reservoir using the concept of symmetry.
The report is outlined as explained in the following paragraphs.
Chapter 1 is a precise literature review to the extent that might be required by the reader to
understand the contents of the next chapters.
Chapter 2 explains how the black oil and compositional models were made using the
correlations given in Chapter 1. The grid model is also discussed in this chapter.
Chapter 3 gives an explanation on the optimization strategy used in the study.
Chapter 4 covers the optimization process and the corresponding results.
In Chapter 5, integrates the conclusions that can be made based on the results obtained in
the previous chapter.
In Chapter 6, outlines the recommendations made based on the conclusions drawn in the
previous chapter. The goal of this chapter is to draw h
areas of research similar to this thesis subject.
Chapter 1: Literature Review
This chapter covers the background information from literature review which will be used in
the discussions in the next chapters.
1.1 Oil and Gas Properties and Correlations

This section gives definitions and correlations for the required PVT properties of oil and
natural gases to make the black oil PVT table for the model under study.
1.1.1 Critical and Reduced Properties

Reduced temperature (Tr) and reduced pressure (pr) are dimensionless properties of a pure
component and are defined as
T
Tr
Tcr
(1.1)
p
pr
.
pcr
(1.2)
Absolute units must be used whilst calculating reduced temperature and pressure. pcr and Tcr
are the true critical properties of a pure component. In most petroleum engineering
applications, pr ranges from 0.02 to 30 for gases and 0.03 to 40 for oils and Tr ranges from <1
to 0.25 for gases and from 0.4 to 1.1 for oils (Whitson and Brule 2000).
1.1.2 Gas Pseudocritical and Pseudoreduced Properties

Average hydrocarbon gas mixture critical properties, known as pseudocritical properties
(ppc and Tpc), can be obtained from mixture specific gravity.
3
Sutton (1985) presented the following correlations for estimation of ppc and Tpc of
hydrocarbon gas mixtures, knowing the hydrocarbon gas mixture specific gravity (g).
Tpc 169.2 349.5 g 74.0 g2
(1.3)
ppc 756.8 131.0 g 3.6 g2
(1.4)
Standing (1981) suggested the following two correlations for dry hydrocarbon gases
(g<0.75).
Tpc 168.0 325.0 g 12.5 g2
(1.5)
ppc 667.0 15.0 g 37.5 g2
(1.6)
The Standing correlations are used in the industry extensively.

Pseudocritical properties of hydrocarbon gas mixtures are used to calculate pseudoreduced
temperature and pseudoreduced pressure for the gas mixture using Eqs. 1.7 and 1.8.
T
Tpr
Tpc
(1.7)
p
ppr
ppc
(1.8)
1.1.3 Gas Deviation Factor

According to the ideal-gas law,
pV nRT ,
(1.9)
where R is the universal gas constant.

At moderate to high pressure or at low temperatures, volume of constituting molecules of a
gas mixture and the corresponding intermolecular forces affect the volumetric behavior of the
gas and the ideal-gas law does not hold anymore. The deviation from ideal-gas law can be
expressed by gas deviation factor, also known as Z-factor. By definition,
volume of one mole of real gas at p and T

volume of one mole of ideal gas at p and T
(1.10)
According to the definition of Z-factor, this parameter is unitless.

Having pseudoreduced pressure and temperature of a given hydrocarbon gas mixture, one
can use Standing-Katz Z-factor chart (Fig. 1) to estimate the Z-factor for that mixture.
Figure 1 - Standing-Katz Z-factor Chart (Whitson and Brule 2000)
Sutton (1985) claimed that his suggested correlations (Eqs. 1.3 and 1.4) are the most
reliable correlations for estimating Z-factor with the Standing-Katz Z-factor chart.
Dranchuk and Abou-Kassem (1975) fitted one of the most accurate equations-of-state to the
Standing-Katz Z-factor chart, which is more convenient for estimating the Z-factor for a gas
mixture. Their equation is expressed as follows:
A
A
A
A
A
A
A
A
5
Z 1 (A1 2 33 44 55 ) pr (A6 7 28 ) pr A9 ( 7 28 ) pr
Tpr Tpr Tpr Tpr
Tpr Tpr
Tpr Tpr
2
pr
2 )(
2 ),
A10 (1 A11 pr
)exp( A11 pr
3
Tpr
(1.11)
where
pr
0.27ppr
(1.11a)
ZTpr
and where the constants A1 to A11 are as follows:

A1 = 0.3265,
A2 = -1.0700,
A 3 = -0.5339,
A 4 = 0.0157,
A 5 = -0.0517,
A6 = 0.5475,
A 7 = -0.7361,
A8 = 0.1844,
A9= 0.1056,
A10= 0.6134,
and A11= 0.7210.
Dranchuk and Abou-Kassem equation can be used for a broad range of pressures and
temperatures and is valid for . ppr
1. Tpr . and for ppr 1
h . Tpr
1.0.
1.1.4 Gas Density

One can derive all volumetric properties of gases from the real gas law. Gas density, g, is
given by
g
pMg
ZRT
(1.12)
Alternatively, in terms of gas specific gravity,

g 28.97
p g
ZRT
(1.13)
in which p is in psia, R is in (psia.ft3)/(oR.lb-mol), T is in oR, and g is in lbm/ft3.
1.1.5 Gas Formation Volume Factor

By definition, gas formation volume factor, Bg, is the ratio of gas volume at specific
temperature and pressure (e.g. reservoir temperature and pressure) to the ideal-gas volume at
standard conditions. Therefore, Bg is given by
Bg
ZTpsc
.
pTsc
(1.14)
1.1.6 Gas Viscosity

Typical reservoir gas viscosities range from 0.01 to 0.03 cp at standard and reservoir
conditions, reaching up to o.1 cp for near critical gas condensate (Whitson and Brule 2000).
Lee-Gonzalez (1966) suggested a gas viscosity correlation (Eq. 1.15) which is nowadays
the most commonly used correlation in PVT laboratories.
A
g A1 x104 exp(A2 g 3 )
(1.15)
where
A1
(9.379 0.01607Mg )T 1.5

209.2 19.26Mg T
(1.15a)
A2 3.448 (986.4 / T ) 0.01009Mg ,
(1.15b)
and
A1 2.447 0.2224 A2 .
(1.15c)
In Lee-Gonzalez gas viscosity correlation, g is in cp, g is in g/cm3, and T is in oR. McCain

(1991) showed that the accuracy of Lee-Gonzalez correlation is 2 to 4% for g <1.
1.1.7 Solution Gas/Oil Ratio

Solution gas/oil ratio, Rs, is the volume of gas at standard conditions librated from a singlephase oil at elevated pressure and temperature (e.g. reservoir pressure and temperature)
divided by the resulting stock-tank oil volume, with units scf/STB (Whitson and Brule
2000).
1.1.8 Bubblepoint Pressure

Petrosky and Farshad (1993) presented a correlation for estimating bubblepoint pressure for
the Gulf of Mexico crude oils. They claimed that their correlation provided improved results
for the Gulf of Mexico oils compared to other correlations developed by other researchers for
California oils, North Sea oils, etc. Their correlation gives
R0.5774
pb 112.727( s0.8439 10 x 12.34) ,
(1016)
where
x 0.00004561T 13911 0.0007916 1.541
API .
(1.16a)
In Petrosky and Farshad correlation, Rsis in scf/STB, T is in oF, and pb is in psia.
1.1.9 Oil Viscosity

Based on the curve fitting of the undersaturated oil viscosity versus pressure data presented
by Beal (1946), Standing (1981) suggested the following correlation for estimation of
undersaturated oil viscosities.
o ob
0.001(p pb )
1.6 0.038 0.56

0.024 ob
ob
(1.17)
Chew and Connally (1959) tried to correlate saturated oil viscosity with dead-oil viscosity.
They proposed the following correlation.
A
ob A1oD2
(1.18)
Different researchers tried to present mathematical relations for A1 and A2 in Chew and
Connally correlation. For example, Standing (1981) suggested the following relations in
terms of Rs for A1 and A2.
8
A1 10
A2
(0.00074Rs 0.00000022Rs2)
0.68
0.0000862Rs
10
0.25
0.0011Rs
10
(18a)
0.062
(18b)
0.00374Rs
10
1.1.10 Oil Compressibility

Petrosky and Farshad (1993) proposed the following correlation for estimation of oil
compressibility of undersaturated oils in the Gulf of Mexico.
0.69357 0.1885 0.3272T 0.6729 p0.5906
co 1.705x107 Rsb
g
API
(1.19)
where 2.464x10-5 < co < 3.507x10-5.

Vasquez and Beggs (1980) also suggested a more general correlation (Eq. 1.20) for
prediction of undersaturated oil compressibility.
co
1433 5Rs 17.2T 1180 g 12.61 o
(1.20)
105 p
1.1.11 Oil Formation Volume Factor

Oil formation volume factor ranges from 1 bbl/STB for oils with little solution gas to about
2.5 bbl/STB for volatile oils (Whitson and Brule 2000).
Petrosky and Farshad (1993) correlation for estimation of bubblepoint oil formation volume
factor for Gulf of Mexico crude oils is
Bob 0.9759 0.00012A1.2
(1.21)
where
A Rs ( g / o )0.5 1.25T .
(1.21a)
One can use oil compressibility and bubblepoint oil formation volume factor to calculate
the variations in oil formation volume factor using Eq. 1.22.
Bo Bob exp[co (pb p)]
(1.22)
1.2 Miscibility
Miscibility is different from solubility. Two fluids are said to be miscible when they combine
in any proportion to form one phase. For example water and methanol are two miscible
fluids. But if a fluid gets dissolved in another fluid up to a certain proportion, it is said to be
soluble in the other fluid. Solubility is a function of the nature of fluids, temperature, and
pressure. An intermediate phenomenon is known as dynamic miscibility where miscibility is
attained gradually, through multi-contact, and involves extraction or vaporization of light
ends (Khatib et al. 1981).
Miscibility can be achieved either by first-contact (absolute) miscibility process or by a
multiple-contact (thermodynamic) miscibility process. First-contact miscibility is a condition
in which the injected gas (solvent) and oil are miscible (i.e. they form a single phase when
mixed in any proportion when first brought into contact at a given pressure and temperature).
Condition of first-contact miscibility for reservoir gasflooding depends on the composition of
the injected gas, composition of oil, temperature, and the injection pressure. By contrast,
fluids that develop miscibility after exchanging components have multiple-contact miscibility
(Oilfield Glossary, First-Contact Miscibility 2013). Multiple-contact (dynamic) miscibility is
a dynamic fluid-mixing process in which the injected gas exchanges components with in
situ oil until the phases achieve a state of miscibility within the mixing zone of the flood
front. In a vaporizing drive, light and intermediate components from the oil phase enter the
gas phase. By contrast, in a condensing drive, intermediate components from the gas phase
enter the oil phase. The process may be a combination of vaporizing and condensing drives
(Oilfield Glossary, Multiple-Contact Miscibility 2013).
1.2.1 Minimum Miscibility Pressure (MMP)

Figure 6 shows a series of hypothetical slimtube experiments. In these experiments, the
solvent displaces oil from the fully oil-saturated slimtube at several pressures. Oil recovery is
shown after 1.2 pore volume (PV) of injection for each pressure. Oil recovery increases with
pressure up to approximately 95 to 98% and then increases very little thereafter. The pressure
at which the break in the recovery curve occurs is called the minimum miscibility pressure
(MMP). If the displacements had been conducted at constant pressure and with increasing
enrichment by components such as ethane, propane, and butane, the break would have been
at the minimum miscibility enrichment (MME). Above the MMP or MME, the displacement
is said to be multiple-contact or dynamically miscible. The increasing recovery with pressure
or solvent enrichment results from in-situ mass transfer of components between solvent and
resident oil. Each pressure increase produces an equilibrium mixture that becomes
compositionally similar at the MMP or MME (Petrowiki, Miscible Flooding 2013).
10
Figure 2 - Slimtube displacements used to determine MMP or MME (Petrowiki, Miscible Flooding
2013)
It should be noted that a slimtube test cannot fully represent the performance in reservoir
rock because it does not account for the effects of factors such as gravity segregation and
reservoir heterogeneity on volumetric sweep (Petrowiki, Miscible Flooding 2013).
CO2 flooding is applicable with medium-gravity oils. At temperatures less than
approximately 125F, MMP is estimated to be as low as 1200 psia. MMP increases with
temperature. (Petrowiki, Designing a Miscible Flood 2013).
1.2.2 Effect of Numerical Dispersion on Miscible Gas Injection

Simulation
Miscible gas injection is one of the most promising EOR techniques which was invented
many years ago and since 1950s it has been implemented in many places around the world.
From a reservoir simulation standpoint, one of the main factors affecting recovery
calculations by simulation of a miscible solvent injection is grid refinement. Numerical
dispersion due to large grid block size, greatly affects oil/gas miscibility in reservoir
simulation of 1D displacement. This, in consequence, results in reduction of local
displacement efficiency. Therefore, refinement of grid cells can diminish the effect of
numerical dispersion in reservoir simulations (Johns et al. 1992; Haajizadeh et al. 1999;
Solano et al. 2001). However, different studies (Stalkup 1990; Lim et al. 1997; Haajizadeh et
al. 2000) on the effect of numerical dispersion on 2D displacement efficiency for miscible
gas flooding drew different conclusions about if very small grid blocks can diminish the error
introduced by numerical dispersion in reservoir simulation of 2D displacement. It seems that
the sensitivity of recovery to grid refinement for miscible flooding is a function of injection
conditions and reservoir descriptions.
11
1.3 SENSOR Reservoir Simulator

SENSOR is a generalized 3D numerical reservoir simulation model for simulating
compositional and black oil problems.
1.3.1 SENSOR Data File Structure

SENSOR data file consists of 3 main sections:
-
Initial Data: The data file starts with TITLE/alphanumeric lines/ENDTITLE, followed
by GRID NX Ny Nz, and remaining Initial Data. The Initial Data ends with the keyword
ENDINIT.
Modification Data: The Modification Data, if present, start with the keyword MODIFY,
following ENDINIT.
Recurrent Data: Recurrent Data normally start with the keyword WELL, and end with
the keyword END, which is the last keyword in the data file.
All the keywords in SENSOR data file are upper case. Apart from the PVT data, there are
almost no differences in the data input for black oil and compositional problems (INITIAL
data formats and specifications for injected gas composition differ). Comments may be
included on any keyword or data line following an !. A comment line is one where the first
h
h

b
b k.
The general layout of SENSOR data file is shown in Appendix 1.
1.3.2 Analytical Relative Permeability and Capillary Pressure Curves in

SENSOR
In research problems or in case of occurrence of large scatter and uncertainty in relative
permeability data, analytical relative permeability (kr) and capillary pressure (pc) data are
often used. Using analytical expressions makes it easy to change kr data for the purpose of
sensitivity runs and history matching.
Endpoint saturations and curvature are the most important properties of kr curves.
KRANALYTICAL keyword in SENSOR is used to represent these properties using the
minimum number of variables.
To use the analytical kr data, the following variables should be entered:
Swc: connate water saturation
Sorw: residual oil saturation to water
Sorg: residual oil saturation to gas
12
Sgc: critical gas saturation

Sgr: trapped gas saturation, optional, default=Sgc
krwro: relative permeability of water at Sw=1-Sorw and Sg=0
krgro: relative permeability of gas at Sw=Swc and So= Sorg
krocw: relative permeability of oil at Sw=Swc and Sg= 0
nw, now, ng, and nog: exponents for analytical kr
The relative permeabilities are calculated by SENSOR as
n
krw krwro [(Sw Swc ) / (1 Sorw Swc )] w
(1.23)
n
krow krocw[(1 Sorw Sw ) / (1 Sorw Swc )] ow
(1.24)
nog
krog krocw[(1 Sorg Swc S g ) / (1 Sorg S wc )]

ng
krg krgro [( S g S gc ) / (1 Sorg Swc S gc )]
(1.25)
(1.26)
Analytical pc are calculated by SENSOR as

A
pcwo A1 A2 (1 S wn ) 3
(1.27)
B
B5
pcwoi B1 B2 (1 S wn ) 3 B4 S wn
(1.28)
pcgo C1 C2 S gn3
(1.29)
where
pcwo: drainage water/oil pc
pcwoi: imbibition water/oil pc
pcgo: drainage gas/oil pc
Swn (Sw Swc ) / (1 Swc )
(1.30)
S gn S g / (1 Swc )
(1.31)
A1 to A3, B1 to B5, and C1 to C3 are variables entered for PCWO, PCWOI, and PCGO
keywords.
13
1.4 Well Productivity Index Calculations

Well index (J) can be calculated using the formulae (Eq. 1.32) presented by Peaceman
(1983). Assume that the wellbore penetrates the center of a grid block in any direction, X, Y,
or Z. The wellbore direction is labeled direction 3 and the directions in the plane
perpendicular to the wellbore are labeled directions 1 and 2. Considering this convention,
block permeability and dimensions in directions 1, 2, and 3 are k1, k2, and k3 and d1, d2, and
d3, respectively. P
for an anisotropic case (k1k2) is

J 0.00708kd3 / (ln(r0 / rw ) s)
(1.32)
where
k k1k2 , k and d are in units of md and ft, and
r0 0.28 [(k2 / k1)0.5 d12 (k1 / k2 )0.5 d22 ]0.5 / [(k2 / k1)0.25 (k1 / k2 )0.25 ]
(1.33)
For an isotropic case (k1=k2), Eq. 1.33 is reduced to

r0 0.14 (h12 h22 )0.5
(1.34)
14
Chapter 2: Model Preparations
This chapter presents the methodology used for making the two fundamental components of
the simulation model: PVT data and grid model. Furthermore, other components of the
model such as well data (productivity index, perforations, well constraints, etc.) are also
discussed in this chapter. The model was made based on the limited data available in open
literature on Wilcox formation in the Gulf of Mexico to assure that it mimics an ultradeep,
high pressure reservoir.
2.1 PVT Data

There are very limited publications on Wilcox formation in the Gulf of Mexico in open
literature and, consequently, one of the major constraints on making the model for this study
was lack of real data. Tables 1 and 2 summarize the only useful data available on Wilcox
formation in open literature which were used as the basis for making the model.
Table 1-Wilcox formation properties found in open literature
15
Table 2-Some other properties of Wilcox formation available in open literature
An initial black oil PVT table was first made using the oil and gas correlations given in
Section 1.1 and the available data in Table 2. The following paragraphs explain step by step
how the initial black oil table was made.
To make the initial black oil table, pseudo-reduced pressure and temperature of the
reservoir fluid were first estimated using the given reservoir temperature and pressure in
Table 2 and Standing correlations (Eqs. 1.5 and 1.6).
Gas deviation factor (Z-factor) was then estimated at pressures between standard pressure
and initial reservoir pressure inclusive using Dranchuk and Abou-Kassem correlation (Eq.
1.11). Having done that, gas density and gas formation factor at each pressure were
calculated using Eqs. 1.12 and 1.14, respectively.
Having gas density calculated, Lee-Gonzalez correlation (Eq. 1.15) was used to estimate
gas viscosity at each pressure.
Wilcox oil is a low-GOR oil. Therefore, solution gas/oil ratio at pressures less than
bubblepoint pressure was obtained by linear interpolation between standard conditions
solution gas/oil ratio (0 scf/STB) and bubblepoint solution gas/oil ratio (181.08 scf/STB).
Having Wilcox oil viscosity at initial reservoir pressure, oil viscosity at bubblepoint
pressure was back-calculated using Beal correlation (Eq. 1.17). The same correlation was
then used to estimate other oil viscosities for undersaturated oil. After that, dead-oil viscosity
was back-calculated using the estimated bubblepoint oil viscosity in the previous step and
Chew and Connally correlation (Eq. 1.18). The same equation was used to estimate other oil
viscosities for saturated oil. A1 and A2 parameters in Eq. 1.18 were obtained from Standing
correlations given in Eqs. 1.18a and 1.18b.
16
Oil compressibility for undersaturated oil was estimated from Petrosky and Farshad
correlation (Eq. 1.19). The estimated undersaturated oil compressibilities were comparable to
estimated values from Vasquez and Beggs correlation (Eq. 1.20).
For estimation of bubblepoint oil formation volume factor, Petrosky and Farshad correlation
(Eq. 1.21) was used. Having done that, linear interpolation was done between standard
conditions oil formation volume factor (1 RB/STB) and bubblepoint oil formation volume
factor. Then, Eq. 1.22 was used to calculate oil formation volume factor at undersaturated
pressures.
Table 3 gives the initial black oil PVT table estimated for Wilcox oil through the
aforementioned procedure.
The black oil PVT data given in Table 3 can be representative of different reservoir fluid
compositions and equation-of-states (EOS). Therefore, it is not possible to estimate a unique
fluid composition and its corresponding EOS from the initial black oil PVT table. On the
other hand, black oil PVT table can only be used in simulation of recovery mechanisms in
which reservoir fluid composition is not altered while for simulation of miscible CO2
injection we needed to make a compositional model. Therefore, it was decided to consider
one of the fluid compositions and EOS which can generate the initial black oil PVT table as
the reservoir fluid composition and EOS for the compositional model.
17
Table 3-Initial black oil PVT table estimated from oil and gas correlations
To estimate a fluid composition and EOS for the reservoir fluid consistent with the initial
black oil PVT table, we tried to tune the fluid composition and EOS parameters given by
Killough and Kossack (1987) in the fifth SPE comparative solution project. The fluid
composition and EOS used in the project is given in Table 4.
18
Table 4-Fluid composition and EOS used in the fifth SPE comparative solution project
As Wilcox oil is a low-GOR oil, the mole fraction of light components in the given fluid
composition in Table 4 was decreased to be more consistent to that of Wilcox formation.
Doing this, a reasonable match was obtained between Wilcox initial bubblepoint pressure
(1160.3 psia) and our fluid initial bubblepoint pressure (1144.4 psia).
Volume shift for all the components were also added to the EOS given in Table 4 to
match the resulting oil gravity to that of Wilcox oil (30.0 oAPI).
To get a match with Wilcox oil viscosity at initial reservoir pressure, critical Z-factors in
the EOS in Table 4 were tuned. However, we did not manage to match the oil viscosities
from the tuned EOS and those estimated in the initial black oil PVT table.
Table 5 gives the final fluid composition and EOS which was used for making the
compositional model in this study.
19
Table 5-Final reservoir fluid composition and equation-of-state used for the model
Table 6 presents the black oil PVT table which is generated from the fluid composition
and EOS given in Table 5 using the BLACKOIL keyword in SENSOR. The data file which
generates this black oil table is presented in Appendix 2. This final black oil table which is
consistent with fluid composition and EOS in Table 5, was used in the black oil runs of this
study.
20
Table 6-Final reservoir black oil PVT table used for the model
Figures 3 to 7 illustrate the comparison between the initial and final black oil PVT table
parameters. The parameters in the initial black oil PVT table were estimated using oil and
gas correlations and the ones in the final black oil PVT table were obtained from the EOS
and fluid composition presented in Table 5.
21
Figure 3-Gas formation volume factor from the initial and final black oil PVT tables
Figure 4-Oil formation volume factor from the initial and final black oil PVT tables: a reasonable
match for initial bubblepoint pressure
22
Figure 5-Oil viscosity factor from the initial and final black oil PVT tables: a perfect match for oil
viscosity at initial reservoir pressure
Figure 6-Solution gas/oil ratio from the initial and final black oil PVT tables
23
Figure 7-Gas viscosity from the initial and final black oil PVT tables
2.2 Grid Model

Another problem in making the model was constraint on the maximum possible number of
grid cells (6000) in SENSOR. This could result in a very course-grid model which, in
consequence, would possibly yield erroneous simulation results due to numerical dispersion.
To reduce the effect of numerical dispersion, one-twentieths of the main cubic reservoir grid
model was selected for the simulation study. To do this the reservoir area for the main cubic
reservoir containing the same volume of oil as initially found in Wilcox formation was
calculated using Eq. 2.1.
OOIP Ah
(1 Swi )
5.615Boi
(2.1)
in which h is reservoir thickness in ft, A is reservoir area in ft2, Boi is in RB/STB, and original
oil in place (OOIP) is in STB.
As shown below, using an average OOIP of 3000 MMSTB and an assumed initial average
water saturation of 0.2 gave a reservoir area of 1.59667108 ft2.
A
3x109 x 1.0616
2
=1.59667x108 (ft )
0.17809 x 140 x 0.2 x (1-0.2)
On the basis of symmetrical cubic reservoir and considering the fact that Wilcox formation
consists of repetitive layers (see Table 1), a cubic section of the reservoir was selected as the
24
final model for the simulation study. The selected section was then discretized to 24x24
uniform numerical grids in XY plane and 10 nonuniform numerical grids in Z direction. Table
7 gives the dimensions of the selected section and the corresponding grid blocks.
Table 7- Dimensions of the selected section of the cubic Wilcox reservoir model
Figure 8 also illustrates the final grid model the dimensions of which are given in Table 7.
As shown in Table 7 and Fig. 8, the final grid model, which was used in this study, is a cubic
model which is one-twentieth of the main cubic reservoir model in volume. It has an area
equal to a quarter of the area of the main reservoir model and a thickness equal to one-fifth of
the thickness of the main reservoir model.
25
Figure 8-Final reservoir section grid model
As shown in Table 8, the 10 grid blocks in Z direction in the final grid model are grouped
in to three layers. The first layer from top is made up of the first three grid blocks, the second
layer consists of the next four grid blocks, and the last three grid blocks make up the last
layer at the bottom. Permeability in each layer is distributed in the constituting grid blocks in
a manner that is consistent with permeability distribution in the individual layers and the
overall average permeability in Wilcox formation as shown in Table 1.
The data used for model equilibration (initialization) are given in Table 9. Due to lack of
data on Wilcox formation, some of these data (water properties, initial water saturation, and
rock compressibility) were taken from available typical data in the literature. Also, a water
depth of 2500 m SSTVD and a reservoir depth of 10,000 m SSTVD for the model were
assumed. These values are within the ranges of those of Wilcox formation in the Gulf of
Mexico.
26
Table 8- Permeability distribution in the final grid model
Table 9- Data used for model initialization
2.3 Relative Permeability and Capillary Pressure Curves

Due to lack of relative permeability data, we had to use analytical kr data. Table 10 gives the
values used for analytical kr calculations.
27
Table 10- Variables used for analytical kr calculations in the model
Figures 9 and 10 illustrate the calculated analytical relative permeability curves by

SENSOR using the variables given in Table 10.
Figure 9-Analytical oil/water relative permeability curves generated by SENSOR
28
Figure 10- Analytical oil/gas relative permeability curves generated by SENSOR
Zero capillary forces were assumed for the model under study.
2.4 Well Data

All layers were perforated in the producer(s) and injector(s) and the default well radius in
SENSOR (0.25 ft) was considered for the wellbores. To be consistent with the typical
bottomhole pressures in Wilcox formation in the Gulf of Mexico, a minimum bottomhole
peressur of 8000 pasi for producer(s) and a maximum injection pressure of 15000 psia for
injector(s) were considered for the wells in the model. Very high production and injection
rates were considered in the model to set the production and injection constratints on the
minimum bottomhole pressure in producer(s) and maximum injection pressure in injector(s).
Well indexes (productivity index and injectivity index) in each layer was calculated using
P
.
Table 11 summarizes the well data used in the final model.
29
Table 11-Well data in the final model
30
Chapter 3: Optimization Strategy
Through this study, we tried to demonstrate how optimization of well placement, inter-well
spacing, and EOR startup time can maximize ultimate and incremental oil productions in an
ultradeep, high pressure oil reservoir.
Ultimate oil recovery resulting from depletion of the ultradeep, high pressure oil reservoir in
this study followed by miscible CO2 injection is a function of numerous parameters which
include but are not limited to injection pressure, BHP, number of wells, injection and
production schemes, inter-well spacing, well placement (well location), CO2 injection startup
time, well completion, etc. Of these parameters, considering the objective of this study,
number of wells, well placement, inter-well spacing, and CO2 startup time were taken as
variables. All other parameters in the model were set as explained in Chapter 2.
The optimization strategy employed in the study consists of two main steps:
1) Determining the maximum ultimate oil recovery (REOR) from pressure depletion of
the reservoir followed by miscible CO2 flooding by finding the optimum CO2
injection startup time (tD) and well spacing: This was done for different well patterns
(2-spot, 3-spot, 4-spot, and 5-spot) to account for different well placements and
number of wells. Therefore, the variables for each well pattern were only tD and well
spacing. We assumed that 1000 STB of oil production was the economic criteria for
end of simulation run (tEOR) for each well pattern scenario.
2) Determining the maximum incremental oil recovery (R) for the optimum tD and
well spacing defined in the previous step: To do this, first, ultimate oil recovery (RD)
from depletion of the reservoir was determined for the optimum well spacing and
over tEOR defined in the previous step. Then, R was determined using Eq. 3.1.
(3.1)
R REOR RD
31
32
Chapter 4: Results and Discussions
4.1 MMP Estimation for CO2 Miscible Injection

In miscible gas injection, injection pressure should be sufficiently more than the MMP
estimated for the reservoir fluid system and injected gas.
The estimated MMP by PhazeComp, which is an EOS-based program, for our reservoir
fluid and CO2 as the injectant was 2499 psia. The estimated MMP is much lower than the
maximum injection pressure of 15000 psia set in the model and, thus, CO2 injection was
miscible throughout the EOR phase. However, we used the first-contact miscibility option in
SENSOR by using the OIL keyword and setting psat equal to zero in the model to make sure
that our displacement mechanism is fully miscible.
The PhazeComp data file used for MMP estimation is given in Appendix 3.
4.2 Well Spacing and tD Optimization for 2-Spot Well Pattern

In the 2-spot well pattern both wells produced for a while and one of the producers (well 2)
was then converted to CO2 injector till end of CO2 injection, where economic limit of 1000
STB of oil production was approached.
Figure 11 shows the plan view of the final grid model. As clear from the figure, the 2-wells
in the model can be placed in thousands of positions with respect to each other and each
position results in a specific cumulative oil production at tEOR for a given tD. To reduce the
number of simulation runs in the simulation process, first, we did a sensitivity analysis on the
effect of moving the wellbore positions from center of the model to the corners and
boundaries of the model on cumulative oil production. To do this, cumulative oil production
after 20 years of pressure depletion of the reservoir for different well spacings
lines shown in Fig. 12 were compared. Similar comparison was done for different
well spacing
h
Fig. 13.
33
Figure 14 shows cumulative oil productions after 20 years of pressure depletion of the
reservoir for all possible well spacings along A
. As clear from the figure,
maximum cumulative oil production was observed along
h
1 was on
i=6 and producer 2 was located on i=18. Furthermore, the maximum cumulative oil
production decreased as we moved the wells from
line. In other terms, as we moved the wells from the center of the model to the boundaries of
the model, maximum cumulative oil production decreased. The same behavior was observed
in pressure depletion of the reservoir followed by CO2 miscible injection.
Figure 15 illustrates cumulative oil productions after 20 years of pressure depletion of the
reservoir for all possible well spacings along
,
, , and
lines. We observed the
maximum cumulative oil production
h
1 was on i=8 and
producer 2 was located on i=17. Furthermore, the maximum cumulative oil production
decreased as we moved h
. In other terms, as we moved the wells from the center of the model to the
corners of the model, maximum cumulative oil production decreased. The same behavior was
observed in pressure depletion of the reservoir followed by CO2 miscible injection.
Figure 11-Plan view of the final grid model
Based on the observations made in the sensitivity analysis (Figs. 14 and 15), to reduce the
number of simulation runs in the optimization process, we went through the two steps of the
optimization process
h
. The optimization
34
process for the 2-spot well pattern along

found the optimum tD
consisting of 1663 simulation runs

Table 12.
Table 12-Optimum well locations and EOR startup time for 2-spot well pattern
Figure 12-Moving well positions to the boundaries of the model
35
Figure 13-Moving well positions to the corners of the model
Figure 14-Cumulative oil productions after 20 years of pressure depletion for
36
Figure 15- Cumulative oil production after 20 years of pressure depletion for different 2-spot well
spacings along
lines
Well 2 was converted to an injector after 5.2 years of production and CO2 injection
continued up to 63.84 years (tEOR) where the economic limit of 1000 STB of oil production
was reached. This optimum tD and well spacing gave a maximum cumulative oil recovery of
44.876 MMSTB which is equivalent to an oil recovery of 29.9% over tEOR.
Figure 16 compares the oil recovery from the optimum tD (5.2 years) and well spacing with
those from the same well spacing but later CO2 injection startup times.
37
Figure 16-Comparison between recoveries from optimum tD and those from other tDs over tEOR for a 2spot well pattern
According to Fig. 17 higher oil production rate is maintained for a longer time in the case of
the optimum tD which in consequence results in maximized cumulative oil production over
tEOR.
38
Figure 17-Effect of EOR startup time on oil production rate and oil recovery
Figs. 18, 19, and 20 demonstrate pressure distribution at optimum tD in grid blocks 1, 2, and
3 of layer 1, respectively. It is clear from these figures that no high pressure isolation region
was left after pressure depletion of these layers regardless of the permeability of these layers.
Similar uniform pressure distribution was observed in all grid blocks of the other 2 layers in
the model regardless of their permeability (see Table 8).
39
Figure 18-Uniform pressure distribution in grid block 1 of layer 1 with k=30 md at optimum tD and for
optimum well spacing
Figure 19- Uniform pressure distribution in grid block 2 of layer 1 with k=19 md at optimum tD and for
40
Figure 20- Uniform pressure distribution in grid block 3 of layer 1with k=0.3 md at optimum tD and for

In the 3-spot well pattern analysis, producer 2 which was converted to an injector at tD was
placed on grid cell (13,12) throughout the simulation run and the other two wells were
relocated through optimization process to find the optimum well spacing and tD.
Similar sensitivity analysis that we did for the 2-spot well pattern was performed for the 3spot well pattern to reduce the number of simulation runs and the same conclusions were
drawn. That is, while keeping producer 2 at grid number (13,12), moving producer 1 and 3
from the center of the model to the boundaries and corners of the model resulted in decrease
in cumulative oil production. h
k h h
consisting of 1099 simulation runs. The found
optimum tD
h h
, are given in Table 13.
Table 13- Optimum well locations and EOR startup time for 3-spot well pattern
Well 2 was converted to CO2 I injector after 3.15 years of production and CO2 injection
continued up to 59.31 years (tEOR) where the economic limit of 1000 STB of oil production
was reached. This optimum tD and well spacing gave a maximum cumulative oil recovery of
57.443 MMSTB which is equivalent to an oil recovery of 38.29% over tEOR.
41
Figure 21 compares the oil recovery from the optimum tD (3.15 years) and well spacing
with those from the same well spacing but later CO2 injection startup times.
Figure 21-Comparison between recoveries from optimum tD and those from other tDs over tEOR for a 3spot well pattern
42
As shown in Fig. 22, higher oil production rate is maintained for a longer time in the case of
the optimum tD which in consequence results in maximized cumulative oil production over
tEOR.
Similar to the case of optimum 2-spot well pattern, for the case of optimum 3-spot well
pattern, uniform pressure distribution was observed at tD in all grid blocks of the 3 layers in
the model regardless of their permeability. This is shown for the 3 grid blocks of layer 1 in
Figs. 23, 24, and 25.
43
Figure 25-Uniform pressure distribution in grid block 3 of layer 1 with k=0.3 md at optimum tD and for

In the 4-spot well pattern, 3 wells (wells 1, 3, and 4) were located on the 3 vertices of a
triangle and the 4th well (well 2) was placed at the center of the triangle. Well 2 was
permanently positioned at the center of the triangle while the other 3 well on the vertices of
44
the triangle were moved uniformly away from well 2 through the optimization process to find
the optimum well spacing for the 4-spot well pattern. Sensitivity to tD was also done for each
4-spot well pattern to find the optimum tD.
Before the optimization process, similar sensitivity analysis to that in the case of 2-spot and
3-spot well patterns was done to investigate the effect of moving the wells in the model to the
boundaries and corners of the model on cumulative oil production. To do this, well 2 at the
center of the 4-spot well pattern wa
,
lines (Fig. 26) and
the cumulative oil production for a 20-year pressure depletion for all the possible 4-spot well
patterns along these lines were obtained and compared (Fig. 27).
Figure 26-Moving well positions to the boundaries and corners of the model
As shown in Fig. 27, moving the wells to the boundaries and corners of the reservoir
resulted in reduction in cumulative oil production. The same behavior was observed for the
case of pressure depletion of the reservoir followed by miscible CO2 injection.
45
Figure 27-Cumulative oil production after 20 years of pressure depletion for different 4-spot well
spacings along
,
lines
Based on the results of the sensitivity analysis, optimization process was performed for all
b
, and
. Table 14 gives the optimum well
spacing and tD found in optimization of the 4-spot well pattern consisting of 525 simulation
runs. These optimum parameters gave maximum cumulative oil production of 62.045
MMSTB over tEOR of 54.11 years which is equivalent to an oil recovery of 41.36%.
Table 14- Optimum well locations and EOR startup time for 4-spot well pattern
Figure 28 compares the oil recovery from the optimum tD (2 years) and well spacing with
those from the same well spacing but later CO2 injection startup times.
Similar to 2-spot and 3 spot well pattern scenarios, as shown in Fig. 29, for the 4-spot well
pattern, higher oil production rate is maintained for a longer time in the case of the optimum
tD which in consequence results in maximized cumulative oil production over tEOR.
46
Figure 28- Comparison between recoveries from optimum tD and those from other tDs over tEOR for a 4spot well pattern
Unlike the case of optimum 2-spot and 3-spot well patterns, for the case of optimum 4-spot
well pattern, we observed non-uniform pressure distribution in the grid blocks of all the 3
47
layers in the model regardless of their permeability. This is shown for the 3 grid blocks of
layer 1 in Figs. 30, 31, and 32. The non-uniform pressure distribution in all layers resulted in
a pressure range of about 12700 to 13800 psia in the model. However, this pressure range
was below the injection pressure of 15000 psia in the model and, hence, no problem in
wellbore injectivity was observed at tD.
Figure 30- Non-uniform pressure distribution in grid block 1 of layer 1with k=30 md at optimum tD and
for optimum well spacing
Figure 31- Non-uniform pressure distribution in grid block 2 of layer 1with k=19 md at optimum tD a d
48
Figure 32- Non-uniform pressure distribution in grid block 3 of layer 1 with k=0.3 md at optimum tD
and for optimum well spacing

In the 5-spot well pattern, 4 wells (wells 1, 3, 4, and 5) were located on the 4 vertices of a
square and the 5th well (well 2) was placed at the center of the square. Well 2 was
permanently positioned at the center of the square while the other 4 well on the vertices of the
square were moved uniformly away from well 2 through the optimization process to find the
optimum well spacing for the 5-spot well pattern. Sensitivity to tD was also done for each 5spot well pattern to find the optimum tD.
To reduce the number of simulation runs, a sensitivity analysis was done similar to those
done for the other well patterns studied earlier. The analysis was done to investigate the effect
of moving the wells in the model to the boundaries and corners of the model on cumulative
oil production. h
b
h
Fig. 33. Well 2 at the center of the 5-spot well patterns was
moved along these lines and cumulative oil production for 20 years of pressure depletion for
all the possible 5-spot patterns along these lines were compared (Fig. 34).
The sensitivity analysis showed that as wells in the 5-spot well pattern get closer to the
boundaries and corners of the model, cumulative production decreases. Similar results to the
ones shown in Fig. 34 were observed for the case of pressure depletion followed by miscible
CO2 flooding.
h
h
.
49
Figure 33- Moving well positions to the boundaries and corners of the model
Figure 34- Cumulative oil production after 20 years of pressure depletion for different 5-spot well
spacings along
,
50
Table 15 gives the optimum well spacing and tD found in optimization of the 5-spot well
pattern consisting of 341 simulation runs. These optimum parameters gave maximum
cumulative oil production of 65.963 MMSTB over tEOR of 48.76 years which is equivalent to
an oil recovery of 43.96%.
Table 15-Optimum well locations and EOR startup time for 5-spot well pattern
Figure 35 compares the oil recovery from the optimum tD (1.64 years) and well spacing
with those from the same well spacing but later CO2 injection startup times.
Similar to other well patterns studied earlier, as shown in Fig. 36, for the 5-spot well
pattern, higher oil production rate is maintained for a longer time in the case of the optimum
tD which in consequence results in maximized cumulative oil production over tEOR.
Figure 35- Comparison between recoveries from optimum tD and those from other tDs over tEOR for a 5spot well pattern
51
Figure 36- Effect of EOR startup time on oil production rate and oil recovery
Like the case of optimum 4-spot well pattern, for the 5-spot well pattern, we observed nonuniform pressure distribution in the grid blocks of all the 3 layers in the model regardless of
their permeability. This is shown for the 3 grid blocks of layer 1 in Figs. 37, 38, and 39. The
non-uniform pressure distribution shows a pressure range of 10900 to 12900 psia in the
model. However, this pressure range is below the injection pressure of 15000 psia in the
model and, hence, no problem in wellbore injectivity was observed at tD.
52
Figure 37-Non-uniform pressure distribution in grid block 1 of layer 1with k=30 md at optimum tD and
Figure 38-Non-uniform pressure distribution in grid block 2 of layer 1 with k=19 md at optimum tD and
53
Figure 39-Non-uniform pressure distribution in grid block 3 of layer 1 with k=0.3 md at optimum tD and
4.6 Effect of Optimized Well Pattern on Cumulative Oil Production and

Incremental Oil Recovery
In Sections 4.2 to 4.5, the effect of optimized well spacing and tD for different well patterns
was discussed. However, no comparison was done for the effect of optimized well patterns on
the cumulative oil production and incremental oil recovery.
Table 16 gives a comparison between the optimization results of different well patterns
studied earlier.
Table 16-Comparison of the optimization results for different well patterns
Figures 40 to 42 compare the effect of each of the discussed optimized well patterns in
terms of well spacing and tD on cumulative oil production and incremental oil recovery for
pressure depletion of the reservoir followed by miscible CO2 injection.
54
According to Figs. 40 to 42, both cumulative oil production and incremental oil recovery
increased with increasing the number of wells in an optimum way.
Figure 40- Effect of optimized well pattern on cumulative oil production
Figure 41-Effect of optimized well pattern on oil recovery
55
Figure 42-Effect of optimized well pattern (number of wells) on incremental oil recovery
56
Chapter 5: Conclusions
A compositional model was made up in this study with the help of limited data available for
Wilcox formation in the Gulf of Mexico to represent an ultradeep, high pressure reservoir.
The model was then used to investigate the effect of number of wells, EOR startup time, well
spacing, and well placement on the incremental and cumulative oil recovery of depletion of
the reservoir followed by miscible CO2 injection. Hundreds and sometimes thousands of
simulation runs were done to find the optimum well spacing and CO2 injection startup time
for each well pattern representing a specified well placement and number of wells. On the
basis of the results and discussions made in Chapter 4, the following conclusions can be
drawn:
1. Miscible CO2 injection strategy could result in considerable incremental oil recovery
in the ultradeep, high pressure reservoir under study.
2. Both incremental oil recovery and cumulative oil recovery were increased with
number of wells on the model.
3. No pressure isolation region was left after primary depletion of the reservoir in any of
the well pattern scenarios and, hence, no problem with injectivity was observed in all
scenarios. This was most probably due to the fact that the reservoir was neither
stratigraphically nor structurally compartmentalized in the model.
4. The higher the number of wells, the earlier the optimized startup time. Therefore, the
optimized EOR startup time is dependent on the well pattern.
5. Sensitivity analysis for well positions with respect to the boundaries and corners of
the model was done before optimization process and it considerably reduced the
number of simulation runs in the optimization process.
6. The optimum location on the model for the injector (well 2) at the center of the 4-spot
and 5-spot well patterns is the center of the model.
57
Chapter 6: Recommendations for Future Research
The following recommendations can be made based on the results discussed in Chapter 4 and
conclusions made in the previous chapter. Some of these recommendations can be used to
conduct further research in the field of ultradeep, high pressure reservoirs.
1. Although miscible CO2 flooding resulted in considerable incremental oil recovery,
this EOR strategy should be evaluated economically and its feasibility should be
studied in terms of CO2 availability.
2. The optimization process in this study did not consider any economic evaluation. In
other terms, the optimization process was conducted only from physical ultimate
recovery perspective. Optimization can also be done to maximize economic ultimate
recovery to take into account the economic aspects of field development.
3. Other EOR strategy such as water flooding and miscible natural gas injection should
be studied and the results should be compared with those of miscible CO2 flooding.
4. We did not consider any aquifer in the model. The effect of an aquifer on the
optimized variables can be studied.
5. No reservoir compartmentalization was considered in the model. To see the effect of
reservoir compartmentalization on pressure distribution just before EOR startup, the
model should be compartmentalized structurally or stratigraphically.
58
References
1. Beal, C. 1946. The Viscosity of Air, Water, Natural Gas, Crude Oil and Its Associated
Gases at Oil Field Temperatures and Pressures. In Trans. of the AIME, Vol. 165,
Issue. 1, 95-115: The American Institute of Mining, Metallurgical, and Petroleum
Engineers Inc. ISSN 0081-1696.
2. Chew, J.N. and Connally, C.A. 1959. A Viscosity Correlation for Gas-Saturated
Crude Oils. In Trans. of the AIME, Vol. 216, 23-25: The American Institute of
Mining, Metallurgical, and Petroleum Engineers Inc.
3. Dranchuk, P.M. and Abou-Kassem, H. 1975. Calculation of Z Factors for Natural
Gases Using Equations of State. J. Cdn. Pet. Tech. 14 (3): 34. PETSOC-75-03-03.
http://dx.doi.org/10.2118/75-03-03.
4. Dusterhoft, R., Strobel, M., and Szatny, M. 2012. An Automated Software Workflow
To Optimize Gulf of Mexico Lower Tertiary Wilcox Sand Reservoirs. Paper SPE
151754 presented at the SPE International Symposium and Exhibition on Formation
Damage
Control,
Lafayette,
Louisiana,
USA,
15-17
February.
http://dx.doi.org/10.2118/151754-MS.
5. Haajizadeh, M., Fayers, F.J., Cockin, A.P. et al. 1999. On the importance of
Dispersion and Heterogeneity in the Compositional Simulation of Miscible Gas
Processes. Paper SPE 57264 presented at the SPE Asia Pacific Improved Oil
Recovery
Conference,
Kuala
Lumpur,
Malaysia,
25-26
October.
http://dx.doi.org/10.2118/57264-MS.
6. Haajizadeh, M., Fayers, F.J., Cockin, A.P. 2000. Effects of Phase Behavior,
Dispersion and Gridding on Sweep Patterns for Nearly Miscible Gas Displacement.
Paper SPE 62995 presented at the SPE Annual Technical Conference and Exhibition,
Dallas, Texas, 1-4 October. http://dx.doi.org/10.2118/62995-MS.
7. Johns, R.T., Fayers, F.J., and Orr, F. 1992. Effect of Gas Enrichment and Dispersion
on Nearly Miscible Displacements in Condensing/Vaporizing Drives. Paper SPE
24938 presented at the 67th SPE Annual Technical Conference and Exhibition,
Washington, DC, 4-7 October. http://dx.doi.org/10.2118/24938-PA.
8. Khatib, A.K., Earlougher, R.C., and Kantar, K. 1981. CO2 Injection as an Immiscible
Application for Enhanced Recovery in Heavy Oil Reservoirs. Paper SPE 9928
presented at the SPE California Regional Meeting, Bakersfield, California, 25-27
March. http://dx.doi.org/10.2118/9928-MS.
9. Killough, J.E. and Kossack, C.A. 1987. Fifth Comparative Solution Project:
Evaluation of Miscible Flood Simulators. Paper SPE 16000 presented at the Ninth
SPE Symposium on Reservoir Simulation, San Antonio, Texas, 1-4 February.
59
10. Lee, A.L., Gonzalez, M.H., and Eakin, B.E. 1966. The viscosity of Natural Gases. In
Transactions of Society of Petroleum Engineers, Vol. 18, Issue 08, 997-1,000. Society
of Petroleum Engineers. ISSN 0149-2136. http://dx.doi.org/10.2118/1340-PA.
11. Lewis, J., Clinch, S., Mayer, D. et al. 2007. Exploration and Appraisal Challenges in
the Gulf of Mexico Deep-Water Wilcox: Part 1-Exploration Overview, Reservoir
Quality, and Seismic Imaging. Presented at the 27th Annual Gulf Coast Section
SEPM Foundation Bob F. Perkins Research Conference, Huston, Texas, USA, 2-5
December.
12. McCain, W.D. Jr. 1991. Reservoir-Fluid Property Correlations-State of the Art. SPE
Res Eng 6(2): 266-272. SPE-18571. http://dx.doi.org/10.2118/18571-PA.
13. Lim, M.T., Pope, G.A., Sepehrnoori, K., et al. 1997. Grid Refinement Study of a
Hydrocarbon Miscible Gas Injection Reservoir. Paper SPE 38060 presented at the
SPE Asia Pacific Oil and Gas Conference, Kuala Lumpur, Malaysia, 14-16 April.
14. Muskat, M. 1949. Physical Principles of Oil Production, first edition, New York:
McGraw-Hill Book Company.
15. Oilfield Glossary. 2014. First-Contact Miscibility
http://www.glossary.oilfield.slb.com/en/Terms/f/first-contact_miscibility.aspx
(accessed 24 May 2014).
16. Oilfield Glossary. 2014. Multiple-Contact Miscibility
http://www.glossary.oilfield.slb.com/en/Terms/m/multiple-contact_miscibility.aspx
(accessed 22 May 2014).
17. Peaceman, D.W. 1983. Interpretation of Well-Block Pressures in Numerical Reservoir
Simulation With Nonsquare Grid Blocks and Anisotropic Permeability. SPE J. 23(3):
531-543. SPE-10528-PA. http://dx.doi.org/10.2118/10528-PA.
18. Petrosky, G.E. and Farshad, F.F. 1993. Pressure-Volume-Temperature Correlations
for Gulf of Mexico Crude Oils. Paper SPE 26644 presented at the SPE Annual
Technical Conference and Exhibition, Houston, Texas, 3-6 October.
19. Petrowiki. 2013. Designing a Miscible Flood (19 August 2013 revision)
http://petrowiki.org/Designing_a_miscible_flood (accessed 25 May 2014).
20. Petrowiki. 2013. Miscible Flooding (19 August 2013 revision)
http://petrowiki.org/Miscible_flooding (accessed 25 May 2014).
21. Rains, D.B., Zarra, L., and Meyer, D. 2007. The Lower Tertiary Wilcox Trend in the
Deep Water Gulf of Mexico. Presented at the AAPG Annual Convention, Long Beach
California, USA, 1-3 April.
22. Solano, R., Lee, S.T., Ballin, P.R. et al. 2001. Evaluation of the Effects of
Heterogeneity, Grid Refinement, and Capillary Pressure on Recovery for MiscibleGas Injection Processes. Paper SPE 71602 presented at the SPE Annual Technical
60
Conference and Exhibition, New Orleans, Louisiana, , 30 September-3 October.

23. Stalkup, F.L. 1990. Effect of Gas enrichment and Numerical Dispersion on EnrichedGas-Drive Predictions. SPE Res Eng 5(4): 647-655. http://dx.doi.org/10.2118/18060PA.
24. Sutton, R.P. 1985. Compressibility Factors for High-Molecular-Weight Reservoir
Gases. Paper SPE 14265 presented at the SPE Annual Technical Conference and
Exhibition, Las Vegas, Nevada, 22-26 September. http://dx.doi.org/10.2118/14265MS.
25. Standing, M.B. 1981. Volumetric and Phase Behavior of Oil Field Hydrocarbon
Systems, Richardson, Texas: Textbook Series, SPE.
26. Whitson, H.C. and Brule, M.R. 2000. Phase Behavior, Vol. 20, 18-46, Richardson,
Texas: Monograph Series, SPE.
27. SENSOR Manual, 2011: Coats Engineering, Inc.
28. Vasquez, M.E. and Beggs, H.D. 1980. Correlations for Fluid Physical Property
Prediction. J. Pet. Tech 32(6): 968-970. SPE-6719. http://dx.doi.org/10.2118/6719PA.
61
Appendix 1 - SENSOR Data File Structure

C
**START OF INITIAL DATA**
TITLE
. . . any number of 132-column alphanumeric lines. . .
ENDTITLE
GRID Nx Ny Nz
C
default solver is RBILU(0)
C
default formulation is IMPES
D4
! if the direct-solution D4 solver is to be used
NF
! if the Nested Factorization solver is to be used
IMPLICIT
! if the fully implicit formulation is to be used
MISC . . . 6 numbers . . .
! water properties, pref
PVTEOS
! if this is a compositional problem
. . . EOS data . . .
PVTBO
! if this is a back oil problem
. . . black oil PVT table . . .
SWT n (SGT n)
. . . tables giving kr and pc vs. saturation . . .
COMPACTABLE
! if compaction is used for rock compressibility
. . . compaction tables . . .
DEPTH VALUE or CON or ZVAR or . . . etc.
. . . array data . . .
THICKNESS VALUE or . . .
KX (and/or TX) VALUE or . . .
KY (and/or TY) VALUE or . . .
KY (and/or TY) VALUE or . . .
POROS (and/or PV) VALUE or . . .
SEP
! surface separator data if a compositional problem
. . . data . . .
INITIAL
. . . data specifying initial P, WOC, GOC, etc. . . .
ENDINIT
! end of Initial Data
C
**MODIFICATION DATA START HERE**
MODIFY PV
! optional, to modify pore volumes
. . . data . . .
MODIFY TX
! optional, to modify transmissibility Tx
. . . data . . .
. . . other optional additional MODIFY data . . .
C
**RECURRENT DATA START HERE**
C
WELL data must precede other well data
WELL
. . . well locations and perforations data . . .
62
C
Remaining data here are order independent
C
Except for chronological order of TIME and data changes
WELLTYPE
. . . defines wells as producers or injectors, and defines units . . .
LIMITWELL
! optional
. . . economic limit data for wells . . .
BHP
! for wells not using Tubing Head Pressure (THP) tables
. . . minimum flowing Bottomhole Pressure (BHP) wells . . .
THP
! for wells using THP tables
. . . minimum wellhead pressures for THP wells . . .
THPTABLE
! needed only if there are THP wells
. . . tubinghead pressure tables for THP wells . . .
RATE
. . . well rate data . . .
WELLPLAT
! if any platforms are to be used
. . . assign wells to platforms . . .
. . . other platform keywords/data . . .
LIMITFIELD . . . 5 numbers . . . ! optional field economic limit data
TIME t1
! proceed to time = t1
TIME t1 dtime
! optional, in place of previous line,
C
proceed to time t1, with printout every dtime days
WELL
! optional, to add new wells or change perforations of old ones
. . . data . . .
RATE
! optional, to change well rates (default rates are shut in, with no recirculation
. . . data . . .
TIME t2
! proceed to time = t2
. . . WELL, RATE, etc. data changes . . .
TIME t3
. . . etc . . . data changes . . . time entries . . .
...
TIME tlast
END
**END OF DATA FILE**
63
Appendix 2 From Compositional to Black Oil

TITLE
Compositional to Black Oil
ENDTITLE
GRID 24 24 10
CPU
MISC 1.0 3.0E-6 62.4 0.7 1.0E-6 5000
KRANALYTICAL
0.2 0.2 0.0 0.0
1.0 1.0 1.0 1.0
2.0 2.0 2.0 2.0
DELX CON
24*267.38
DELY CON
24*267.38
THICKNESS ZVAR
3*6.82 4*7.535 3*29.8
POROS CON
0.2
DEPTH CON
24606
KX ZVAR
3*20.40 4*6.12 3*1.10
KY EQUALS KX
KZ EQUALS KX
MOD
1 24 1 24 1 10 * 0.1
BLACKOIL 1 15 31 EXTEND
PRESSURES 14.7 250 500 750 1000 1160.30 1500 2000 2250 2500 2750 2800 3000 4000 5000 6000
7000 8000 9000 10000 11000 12000 13000 14000 15000 16000 17000 18000 19000 20000 20300
RESERVOIR FLUID
0.000000 0.000000 0.280743 0.043155 0.100696 0.287703 0.215777 0.071926
INJECTION GAS
1 0 0 0 0 0 0 0
ENDBLACKOIL
PVTEOS
200
CPT
PC
TC
MW
AC
ZCRIT
SHIFT
OMEGA OMEGB
PCHOR
CO2 1069.51
547.42
44.010 0.2250 0.27433 0.00189 0.45724 0.0778
80.0
N2 491.680
227.29
28.000 0.0400 0.29178 -0.16453 0.45724 0.0778
59.1
C1 667.800
343.00
16.040 0.0130 0.29345 -0.15193 0.45724 0.0778
71.0
C3 616.300
665.70
44.100 0.1524 0.27630 -0.06428 0.45724 0 .0778
151.0
C6 436.900
913.40
86.180 0.3007 0.25862 0.09681 0.45724 0.0778
241.8
C10 304.000 1111.80 142.290 0.4885 0.23435 0.20911 0.45724 0.0788
376.5
C15 200.000 1270.00 206.000 0.6500 0.19872 0.40916 0.45724 0.0778
529.4
C20 162.0000 1380.0000 282.0000 .8500000 0.20143 0.40155 .4572400 .0788
711.80
BIN
64
-0.02000 0.10000 0.13500 0.10000 0.10000 0.10000 0.10000

0.03600 0.08000 0.10020 0.10000 0.10000 0.10000
0.00000 0.00000 0.00000 0.05000 0.05000
0.00000 0.00000 0.00500 0.00500
0.00000 0.00000 0.00000
0.00000 0.00000
0.00000
PSM
INITIAL
DEPTH PSATBP
24606 1160.3
PINIT 20300
ZINIT 24606
ENDINIT
WELL
I
J K PI
WELL1
17 8 1 0.413
17 8 2 0.413
17 8 3 0.413
17 8 4 0.137
17 8 5 0.137
17 8 6 0.137
17 8 7 0.137
17 8 8 0.097
17 8 9 0.097
17 8 10 0.097
WELL2
8 17 1 0.413
8 17 2 0.413
8 17 3 0.413
8 17 4 0.137
8 17 5 0.137
8 17 6 0.137
8 17 7 0.137
8 17 8 0.097
8 17 9 0.097
8 17 10 0.097
WELLTYPE
WELL1 STBOIL
WELL2 MCFINJ
INJGAS
WELL2
1 0 0 0 0 0 0 0
BHP
WELL1 5000
WELL2 15000
65
RATE
WELL1 1.E10
WELL2 1.E10
TIME 7300 100
END
66
Appendix 3 PhazeComp Data File for MMP Estimation

TITLE "MMP Calculations"
TEST1 ON
TEST2 ON
CHAR
EOS PR
Component
CO2
N2
C1
C3
C6
C10
C15
C20
MW
Tc(R)
44.01
547.420
28.00
227.290
16.040 343.000
44.100 665.700
86.180 913.400
142.29 1111.80
206.00 1270.00
282.00 1380.00
Binaries CO2 N2
CO2
-0.02
N2
C1
C3
C6
C10
C15
C20
END
C1
0.10
0.036
Pc(psia)
1069.51
491.68
667.80
616.30
436.90
304.00
200.00
162.00
C3
0.135
0.08
0
AF
0.2250
0.0400
0.0130
0.1524
0.3007
0.4885
0.6500
0.8500
C6
C10
0.1
0.1
0.1002 0.1
0
0
0
0
0
C15
C20
0.1
0.1
0.1
0.1
0.05 0.05
0.005 0.005
0
0
0
0
0
0
MIX OIL 0.0 0.0 0.280743 0.043155 0.100696 0.287703 0.215777 0.071926
MIX INJE 1 0
0 0 0 0
0
0
MIX FEED OIL 1 MOLE
TEMP 200 F
PRESS 0 PSIA
MIX FEED OIL 1 MOLE
TEMP 200 F
PRES PSIA
PSAT
MMP STAGES 500
EOF
67
Appendix 4 Submitted Paper Based on the Thesis Work

A paper was submitted to the 19th Middles East Oil & Gas Show Conference and is still pending for approval
by the review committee. The same paper will be submitted to similar EAGE and SPE events.
68

Optimization of Well Spacing and CO2

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Optimization of Well Spacing and CO2

Uploaded by

Copyright:

Available Formats

Optimization of Well Spacing and CO2

Miscible Flooding Startup Time in an

Ahmad Hossein Zadeh

Norwegian University of Science and Technology